The Journal of arthroplasty | 2021
What is the Clinical Utility of Synovial Alpha Defensin Testing of Antibiotic Spacers Before Reimplantation?
BACKGROUND\nThe purpose of this study was to evaluate the diagnostic performance of standalone alpha defensin (AD) testing of antibiotic spacers during two stage exchange and to determine if the addition of AD testing to other commonly used laboratory tests improves the ability to detect persistent infection in an antibiotic spacer.\n\n\nMETHODS\nCases of two-stage exchange for periprosthetic joint infection from 2016 to 2019 at a single institution were retrospectively reviewed. Cases were classified as persistently infected or not infected in accordance with 2014 and 2018 Musculoskeletal Infection Society criteria to determine if AD provided any clinical utility beyond the other commonly used tests that make up both criteria. Delphi Consensus criteria at 1 year were used as the gold standard for determining recurrent periprosthetic joint infection.\n\n\nRESULTS\nFifty-two spacers (25 hips and 27 knees) in 51 patients were included for analysis. Five spacers were persistently infected based on Musculoskeletal Infection Society criteria. One spacer underwent reresection and the remaining 4 underwent reimplantation with no subsequent infectious complications. All 48 patients who were categorized as not infected underwent reimplantation; at 1 year postoperatively, 7 (13%) had failed due to infection. Three spacers (6%) had a positive AD test. Two spacers with positive AD tests underwent reimplantation, neither had failed at 1 year postoperatively. Sensitivity of standalone AD testing was 0%, and specificity was 96%.\n\n\nCONCLUSION\nStandalone AD testing for the purpose of predicting repeat infection after two-stage exchange arthroplasty exhibits sensitivity of 0% and low predictive value. Addition of synovial AD testing did not increase the diagnostic performance of commonly used synovial and serologic markers of infection.\n\n\nLEVEL OF EVIDENCE\nIV-retrospective cohort study.